llvm dev - Feb 2017 - [cfe-dev] [4.0.0 Release] Release Candidate 2 source and binaries available

> Please try it out, run tests, build your favourite projects and file
> bugs about anything that needs to be fixed, marking them as blockers
> of http://llvm.org/pr31622.
I have encountered very long compile times for three large source files
containing generated/unrolled code at -O1.
We are talking about 10+ hours here without completing, so it looks very much
like an endless loop.
The processes are using 15, 22 and 27 GB of memory but do not appear to grow
further.
This worked fine in the past, so appears to be a regression.

These stacks seem to be prevalent:

#0  0x0000000001850588 in llvm::SUnit::addPred(llvm::SDep const&, bool) ()
#1  0x0000000001855265 in
llvm::ScheduleDAGInstrs::addChainDependency(llvm::SUnit*, llvm::SUnit*, unsigned
int) ()
#2  0x0000000001855519 in
llvm::ScheduleDAGInstrs::addChainDependencies(llvm::SUnit*,
llvm::ScheduleDAGInstrs::Value2SUsMap&) ()
#3  0x000000000185f3ba in
llvm::ScheduleDAGInstrs::buildSchedGraph(llvm::AAResults*,
llvm::RegPressureTracker*, llvm::PressureDiffs*, llvm::LiveIntervals*, bool) ()
#4  0x00000000017d3c89 in llvm::ScheduleDAGMILive::buildDAGWithRegPressure() ()
#5  0x00000000017d3d13 in llvm::ScheduleDAGMILive::schedule() ()
#6  0x00000000017d015e in (anonymous
namespace)::MachineSchedulerBase::scheduleRegions(llvm::ScheduleDAGInstrs&,
bool) [clone .isra.553] ()
#7  0x00000000017d2c11 in (anonymous
namespace)::MachineScheduler::runOnMachineFunction(llvm::MachineFunction&)
()
#8  0x000000000177ca13 in
llvm::MachineFunctionPass::runOnFunction(llvm::Function&) ()
#9  0x0000000001a29983 in
llvm::FPPassManager::runOnFunction(llvm::Function&) ()
#10 0x0000000001a29a2c in llvm::FPPassManager::runOnModule(llvm::Module&) ()
#11 0x0000000001a2a28f in llvm::legacy::PassManagerImpl::run(llvm::Module&)
()

With these mixed in from time to time:

#0  0x0000000001850825 in llvm::SUnit::ComputeDepth() ()
#1  0x00000000017d22a4 in llvm::GenericScheduler::registerRoots() ()
#2  0x00000000017c9002 in
llvm::ScheduleDAGMI::initQueues(llvm::ArrayRef<llvm::SUnit*>,
llvm::ArrayRef<llvm::SUnit*>) ()
#3  0x00000000017c9439 in
llvm::ScheduleDAGMILive::initQueues(llvm::ArrayRef<llvm::SUnit*>,
llvm::ArrayRef<llvm::SUnit*>) ()
#4  0x00000000017d3de3 in llvm::ScheduleDAGMILive::schedule() ()
#5  0x00000000017d015e in (anonymous
namespace)::MachineSchedulerBase::scheduleRegions(llvm::ScheduleDAGInstrs&,
bool) [clone .isra.553] ()
#6  0x00000000017d2c11 in (anonymous
namespace)::MachineScheduler::runOnMachineFunction(llvm::MachineFunction&)
()
#7  0x000000000177ca13 in
llvm::MachineFunctionPass::runOnFunction(llvm::Function&) ()
#8  0x0000000001a29983 in
llvm::FPPassManager::runOnFunction(llvm::Function&) ()
#9  0x0000000001a29a2c in llvm::FPPassManager::runOnModule(llvm::Module&) ()
#10 0x0000000001a2a28f in llvm::legacy::PassManagerImpl::run(llvm::Module&)
()


Are there any new optimization passes I could try switching off?
I could not find any mention in the release notes of new passes.

Best regards,
Martin

On 15 February 2017 at 10:24, Richtarsky, Martin via llvm-dev
<llvm-dev at lists.llvm.org> wrote:
> I have encountered very long compile times for three large source files
containing generated/unrolled code at -O1.
> We are talking about 10+ hours here without completing, so it looks very
much like an endless loop.
> The processes are using 15, 22 and 27 GB of memory but do not appear to
grow further.
> This worked fine in the past, so appears to be a regression.
Ouch!

> Are there any new optimization passes I could try switching off?
> I could not find any mention in the release notes of new passes.
It doesn't seem to be any new passes, but the scheduler. Did this work
in RC1? If so, it'll be a lot easier to identify the cause. If not, a
bisection might help.

cheers,
--renato

This may be a shot in the dark (I don't even know whether your revision
includes the culprit revision), but the symptoms seem similar to:
https://bugs.llvm.org//show_bug.cgi?id=31890

On Wed, Feb 15, 2017 at 12:43 PM, Renato Golin via llvm-dev <
llvm-dev at lists.llvm.org> wrote:
> On 15 February 2017 at 10:24, Richtarsky, Martin via llvm-dev
> <llvm-dev at lists.llvm.org> wrote:
> > I have encountered very long compile times for three large source
files
> containing generated/unrolled code at -O1.
> > We are talking about 10+ hours here without completing, so it looks
very
> much like an endless loop.
> > The processes are using 15, 22 and 27 GB of memory but do not appear
to
> grow further.
> > This worked fine in the past, so appears to be a regression.
>
> Ouch!
>
>
> > Are there any new optimization passes I could try switching off?
> > I could not find any mention in the release notes of new passes.
>
> It doesn't seem to be any new passes, but the scheduler. Did this work
> in RC1? If so, it'll be a lot easier to identify the cause. If not, a
> bisection might help.
>
> cheers,
> --renato
> _______________________________________________
> LLVM Developers mailing list
> llvm-dev at lists.llvm.org
> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20170215/229bcece/attachment.html>